Following the glorious victory of the civil courts over the Westboro Baptist Church, I decided to see if there was any ranting about the court’s verdict on the site.
Amazingly, I had forgotten the URL “godhatesfags”, so I googled “Westboro Baptist Church”. I was surprised by the Google description:
Surely, surely, nobody from Westboro actually wrote that, did they?
I know Google have done some behind-the-scenes things before (e.g. “French Military Victories”) - is this another example?
Ah. They’re getting the description from the Open Directory Project, which is human-edited, but not necessarily - in fact probably most often not - by the owner of the linked site.
Google had nothing to do with this. Someone just spoofed the layout of Google. Try it now, and note that the graphics are in different places when you use search vs. “I’m feeling lucky.”
Each submitted site has various details entered by the person who submits the site to DMOZ.org, who has to provide a return email address. DMOZ is meant to have human intervention to approve things. But I guess some things slip through. OR, the human intervention occurred at the editorial stage within DMOZ itself. It never occurred to me that you could submit any site, not just your own. D’oh.
Mind you, if Phred had a proper META DESCRIPTION tag on his putrid front page, then he’d avoid this. But I’m not going to tell him.
I’d always assumed that the “summary” on the Google page comes from META tags on the site itself, but from this thread it appears that it doesn’t. Or is it only for certain sites that Google uses the ODP?
In the old days, at stage 3, this meant that the description would sometimes say “Home | About | News” etc. where the navigation was built from text. However, I think Google’s got a bit smarter and now identifies the first ‘real’ text on the page.